This album stands as a monument to a particular space, place and time: the Chapel Royal of Linlithgow Palace, as it once stood, at the turn of the sixteenth century. Now a ruined shell, with no roof or windows, clinging to the side of the peel above Linlithgow Loch, it was once the great pleasure palace of the kings and queens of Scotland, and the birthplace of James V and Mary, Queen of Scots. As a refuge for the royal family between the bustle of the capital, Edinburgh, and the main royal residence in Stirling, the building once resounded to music sung by the skilled musicians of the itinerant chapel royal, surrounded by magnificent decorations and sculptures. Almost none of this—the music or the building, save its walls—survives.

What we do have, though, is the famous Carver Choirbook, from which all the polyphonic works on this recording, with one exception, are taken. This is one of only two large-scale collections of music to survive from pre-Reformation Scotland, the long-term work of its principal scribe, the Augustinian canon Robert Carver (also known as Robert Arnot), whose name appears in a number of entries in the source. Whether the collection itself had a role to play at Linlithgow is unknown; clearly, however, it was compiled for a sophisticated chapel, almost certainly a royal one. Eight of its twenty-seven works were written by Carver himself in the first half of the sixteenth century, but we have not chosen any of them for this album. Our reconstruction of the palace focuses instead on a slightly earlier period, before the building of an organ within the chapel and the consequent changes to its internal layout. The centrepiece of our recording is a magnificent Mass cycle, found within a layer of the choirbook containing—alongside Dufay’s Missa L’homme armé—works both anonymous and from the mid-to-late fifteenth century. This, along with a companion cycle, has previously been described as either Continental or English, but we now believe the pair to be the oldest surviving Mass cycles of Scottish origin.

Given the known proclivities of Linlithgow’s overlord, King James IV, the present cycle may well have found a place in the king’s devotions. Saint Katherine seems to have held a special place in his observances, as we will see below. It is perhaps no surprise, therefore, that the famous tale of James being warned of his impending death at Flodden by a spectre takes place in the St Katherine aisle of Linlithgow Parish Church. That the warning came from the saint is rendered the more likely given her particular reputation as an intercessor. This derives from her imploration to Christ at the moment of her own death (in the words of The Golden Legend): ‘I beg of you that anyone … who invokes me at the moment of death or in any need may receive the benefit of your kindness.’ We know that James celebrated Masses for Saint Katherine in the chapel at Linlithgow Palace, having given significant funds for the celebration of her feast there in both 1490 and 1497.

We know also from surviving records that James more generally had ‘chapele geir’ and ‘organis’ in the royal chapel at Linlithgow, originally transporting these as necessary from the other royal chapels, in Stirling and Edinburgh, but eventually having them permanently installed. The king spent many important occasions there, and was present particularly often for Easter, including—as a sixteen-year-old—in 1489, and again in the 1490s, most probably for the first official use of the new chapel; his last Eastertide visit occurred in 1512, the year prior to his death on the battlefield at Flodden, and perhaps the occasion for his putative spectral warning. The treasurer’s accounts show that he often spent Yuletide in Linlithgow too, seemingly visiting the palace frequently during the most important occasions in the Christian calendar.

The music
The music recorded here is dedicated to the veneration of Christianity’s two foremost female intercessors. The first of these is clearly the Virgin Mary—more on this below—but the second, Saint Katherine, is a virgin martyr who yields in weight of medieval veneration only to Christ’s mother. King James IV’s particular veneration of Katherine centred on a healing well in the village of Liberton, just outside Edinburgh, which he visited no fewer than fifteen times between 1502 and 1512. The legendary origins of the well have been traced to James’s ancestor Queen Margaret, consort to King Malcolm. The story goes that the queen had been presented, by a pilgrim returning from the Holy Land, with a vial of healing oil from Saint Katherine’s shrine at Sinai. On accidentally dropping the vial, the queen was amazed to see the emergence of a spring of black, oily substance. The viscous substance came to be seen to have healing powers, capable of curing rashes and scabies.

Missa Horrendo subdenda rotarum machinamento—the ‘Catherine Wheel Mass’, as we have taken to calling it—has only recently had its structural chant identified as belonging to a responsory for Saint Katherine of Alexandria. The work is both strange and beautiful; its underlying structure mirrors that of contemporary Mass cycles from south of the border, but its surface details, like those of its companion, the Missa Rex virginum, are quite unlike anything else from the period. It makes frequent use of the octave-leap cadence, which would seem to be a rather antiquated feature for a Mass of c1460, and of a particular kind of floridity that seems on the one hand redolent of the slightly later Eton repertory, and on the other quite distinctive. Its opening movement is missing two folios, robbing us of half of the voices of its opening movement, which we have refashioned in a reconstruction for this recording.

We place the Mass here in the context of a series of snapshots, as it were, from a succession of alleged Saint Katherine’s day devotions at Linlithgow. Our proceedings begin with the extraordinary Matins responsory itself—its bewildering musical machinations vividly recalling those of the famous wheel on which the saint was tortured—which, quoted in the tenor, forms the basis of the Mass. The Sarum introit for Saint Katherine, Dilexisti iustitiam, serves to usher in the glorious polyphonic Mass cycle itself, its movements heard without interruption. Following this, we fast-forward to Vespers, with one of the Carver Choirbook’s elaborate and brilliant settings of the Magnificat, alternating its polyphonic verses with chant set in a four-voice formulaic succession of intervals following the ‘fourth kind of faburdoun’ discussed in the mid-sixteenth-century treatise on the ‘Art of Music’ and known as the ‘Scottish Anonymous’. We end proceedings, as every liturgical day would have ended, with a Marian antiphon following Compline, calling in this instance on the famous Ave Maria, mater Dei by William Cornysh found in the Eton Choirbook. In doing this, we invite comparison with music from that great collection with which our Carver selections stand in oblique comparison. At the same time, moreover, we invoke the English repertory whose representatives also found their way into the Carver book, the expression, perhaps, of Anglophile influx brought with Margaret Tudor, the elder daughter of Henry VII of England and wife of James IV from 1503 until his untimely demise a decade later.

Andrew Kirkman © 2021

This recording is one of several products of two projects funded by the Arts and Humanities Research Council—‘Space, place, sound and memory: Immersive experiences of the past’ and ‘Hearing historic Scotland’—which have sought to bring back to life the lost performance space of Linlithgow Chapel. The projects have set out to reconstruct architecture, interior decoration, and acoustics in immersive virtual reality, and then to situate appropriate music within this performance context. All of the music contained on this recording is situated within this reconstructed acoustic, one markedly more intimate and exposed than one might expect for a recording of this type, or indeed than conventionally experienced in concert performance. We find this a particularly compelling presentation for this repertoire since it allows the listener to hear the finer details of the counterpoint and especially the decorative filigree of the upper voices. Importantly, we see this acoustic as approaching that in which much of this repertoire must originally have been heard. Music designed for private chapels would, of course, have often been performed in smaller spaces, decorated with tapestries, drapes and soft furnishings—an acoustic environment which is quite distinct from that experienced in the bare stone surroundings of modern cathedrals and churches.

You can enjoy the full virtual-reality re-enactment on site at Linlithgow Palace and at St Cecilia’s Hall, Edinburgh. This allows the listener-viewer to walk through the reconstructed chapel for a full, multi-sensory, immersive experience. The album itself offers an experience quite unlike any other: the chance to enjoy a number of works in as accurate a reconstruction as we were able to achieve of a space in which we believe they may once have been heard. If you would like to investigate the difference that changing acoustics can make on performance, please explore the downloadable app (available via the QR code in the accompanying booklet). This allows you to compare the music as it would sound performed in the reconstruction versus its sound in the chapel as it stands today, with its missing ceiling and windows, as well as an opportunity to view our physical reconstruction of the chapel from the perspective of the listener to this recording.

The virtual-reality reconstruction
Our reconstructions started life as highly detailed LIDAR scans of the site, produced via a rotating laser gun that takes measurements of the building, accurate to a fraction of a millimetre, allowing for a detailed model of the structure as it currently stands. This model is constructed from hundreds of thousands of polygons, giving a degree of detail which is both unnecessary and too complex for the VR system to process.

To overcome this, we vastly reduced the number of polygons while maintaining the accurate structural details required for modelling. LIDAR also captures high-resolution images which can be overlaid on the structural model to give it the required texture. We have lost none of the details of the space, as the textures are multi-layered, allowing us to see all the nooks and crannies of the building today.

All of this gives us a detailed re-creation of the building as it currently stands, but it does not give us a sense of what the building might have looked like in the past. To do this, we used the detailed measurements as a structural basis, and started to reconstruct the building around it. Working with archaeologists of standing buildings, historians and art historians, as well as the archival records, we were able to reconstruct the layers of building work on the chapel, initially rebuilding its roof and supporting beams. Glass was added to the windows, along with the rising ledges beneath them. Plaster was added to the walls, with accompanying trompe-l’œil painted tiles, as seen in Stirling Palace. Based on examples found in archaeological digs at the palace, clay tiles were added to the floor. There is no evidence of doors ever having been attached to the doorways at either end of the chapel, since there is nowhere for the attachments to have fitted, so both entries to the chapel are instead closed with drapes. We have also returned devotional sculptures to the now empty niches, selecting representations of appropriate saints.

The acoustics
Once we had a sense of how the interior of the building might once have looked, we were able to begin our acoustic reconstruction. A study of the acoustic properties of the materials used to construct the building was carried out. Materials reflect and absorb sound to different degrees at different frequencies. An object with a rough surface will also scatter the reflected sound more than one with a smooth surface—an additional factor to be considered. All surfaces in our reconstruction had to be tagged with this data so that we could calculate the impulse response of the space.

An impulse response is a measure of all the reflections at all audible frequencies between the sound source and receiver. A process known as ray tracing was employed, whereby reflections from all the surfaces and objects in the space were simulated to produce an impulse response. From there, we could imprint the acoustic characteristics of the modelled space on our recorded music. Our listener (in this case, you) is situated relatively centrally in the chapel. The dimensions of the chapel make for a very different experience from that which the listener would have in most modern settings. The chapel is only roughly eight metres wide and yet is fifteen metres long and eleven metres high—in other words, long, thin and very tall. The lack of width, combined with the positioning of the seating for the royal party, also makes the positioning of the singers around a lectern a challenge, especially when trying to leave space for the appropriate liturgical events to happen. Perhaps the most important point, at least concerning our aural experience, is that the singers would not necessarily have been facing the ‘audience’, as one might expect in a concert setting; a significant portion of the sound in our reconstruction is consequently directed towards the drapes, with the altarpiece acting as a resonator.

The recording process
In order to produce as clean a reconstruction of the acoustics as possible, it was important to record the music in a setting which had close to no natural acoustic. We therefore recorded within an anechoic chamber, the kind of space more usually used by engineers and physicists to test materials. This allowed us to produce a performance as though it were taking place in our reconstructed space, overlaying the reconstructed acoustic of Linlithgow Palace Chapel without interference from the acoustic of a studio or other venue. Anechoic chambers do not make the most natural performance spaces. They are claustrophobic, hot, often (as in this case) with springy floors, and they offer very little in the way of feedback to the performer. On the other hand, they do make for very close aural focus between the performers, and hence facilitate exceptional uniformity of tone and vowel production. Nonetheless, it is also important to ensure that the performance decisions that are taken reflect the intended reconstructed acoustic, rather than the ‘dead’ acoustic found in the chamber. Our entire production process was therefore significantly changed for this recording.

The Binchois Consort would like to thank James Cook for editions and completions of the music, Henry Howard for assistance with the translation of Horrendo subdenda rotarum machinamento, and Andrew Chadwick, James Cook and Rod Selfridge for assistance in preparing the anechoic chamber. We would also like to thank the AHRC for generously funding both the album and the project on which it was based, and Historic Environment Scotland for access to Linlithgow Palace.

Andrew Kirkman © 2021